Smell Is Coded in Grammar and Frequent in Discourse: Cha’Palaa Olfactory Language in Cross-Linguistic Perspective
Total Page：16
File Type：pdf, Size：1020Kb
■ Simeon Floyd ■ Lila San Roque Universidad San Francisco de Quito Radboud University Nijmegen sfl[email protected] Max Planck Institute for Psycholinguistics [email protected] ■ Asifa Majid Radboud University Nijmegen Max Planck Institute for Psycholinguistics [email protected] Smell Is Coded in Grammar and Frequent in Discourse: Cha’palaa Olfactory Language in Cross-Linguistic Perspective Abstract It has long been claimed that there is no lexical field of smell, and that smell is of too little validity to be expressed in grammar. We demonstrate both claims are false. The Cha’palaa language (Ecuador) has at least 15 abstract smell terms, each of which is formed using a type of classifier previously thought not to exist. Moreover, using conversational corpora we show that Cha’palaa speakers also talk about smell more than Imbabura Quechua and English speakers. Together, this shows how language and social interaction may jointly reflect distinct cultural orientations towards sensory experience in general and olfaction in particular. [ol- faction, sensory anthropology, Cha’palaa, Imbabura Quechua, English] Introduction: Taking Account of Diversity in the Language of the Senses he language of the senses is at the heart of longstanding debates about the degree to which language may influence or reflect differences in how people Tacross cultures perceptually approach the world. The language of vision, and particularly of color, has provided a contentious example of how languages are claimed to develop in similar ways, based on universal cognitive and perceptual principals (Berlin and Kay 1969; Kay and Regier 2003; Kay, Berlin, and Maffi 2011). These generalizations have been problematized by Lucy (1997) for, among other things, not taking into account the specifics of individual languages, citing examples like the Hanunoo language of the Philippines in which “color” terms imply more than just color (as described in Conklin 1955; see also Wierzbicka 2005). Additional studies of minority languages like Yel^ı Dnye of Papua New Guinea (Levinson 2000) or Candoshi of Amazonian Peru (Surralles 2016) have raised further questions about Journal of Linguistic Anthropology, Vol. 28, Issue 2, pp. 175–196, ISSN 1055-1360, EISSN 1548-1395. Copyright © 2018 American Anthropological Association. DOI: 10.1111/jola.12190. 175 176 Journal of Linguistic Anthropology the universality of claims about color terms in the face of cultural diversity (see Majid 2015a for further discussion). In the domain of olfaction, universalist proposals are also beginning to be challenged by studies of diverse, lesser-known languages from around the world. Claims that smell is universally impossible to describe abstractly (Sperber 1975:115– 16; Olofsson and Gottfried 2015; Yeshurun and Sobel 2010), and the related implication that olfactory language will be universally minimal across cultures (Lawless and Engen 1977; Stoddart 1990; Ackerman 1991; Wilson and Stevenson 2006; Olofsson and Gottfried 2015) are being confronted by research on the Aslian languages of the Malay Peninsula (Burenhult and Majid 2011; Tufvesson 2011; Wnuk and Majid 2014; Majid and Burenhult 2014; Majid and Kruspe 2018), among other languages with extensive abstract smell lexicons from northern Mexico (O’Meara and Majid 2016) to Africa (e.g., van Beek 1992; Blench and Longtau 1995; Hombert 1992; Storch and Vossen 2007), and Amazonian South America (Shepard 1999). These findings suggest—as has long been argued in sensory anthropology (e.g., Stoller 1989; Howes 1991; Classen 1997; Howes 2003)—that there is not a universal, biologically determined orientation to perceptual experience, but rather that sensory cultures are diverse (or feature different sensory “models”; Classen 1997).1 This diversity suggests that the marginalization of smell may instead be a quirk of some Western cultures (Classen, Howes, and Synnott 1994; Majid 2015b; see also McGann 2017; on Western bias in selection of research subjects more generally, see Henrich, Heine, and Norenzayan 2010), related to historical transformations of the olfactory environments of Western urban contexts in connection to ideologies of hygiene and sanitation (see Corbin 1986 on France). People living outside such contexts may show greater sensitivity to smell, as one study comparing urban Germans to the Tsimane’ of the Bolivian Amazon has shown (Sorokowska et al. 2013). Is the marginalization of smell actually a Western-specific cultural trait that can vary in different societies? The role of language has not always been taken into account in studies of olfaction, even as the categories found in language provide a key entry point into how different cultural models approach sensory experience. Sensory anthropologists have critiqued how the “linguistic turn” in anthropology drew attention to the visual medium of the “text” at the expense other kinds of sensory experience (Howes 2005), but—as Majid and Levinson (2011) point out—focusing on indigenous linguistic expressions in diverse languages and cultures has precisely the opposite effect: it draws our attention to the diversity of cultural categories that partition the sensory domains in different ways (see also the call for attention to the linguistics of sensory language in Lucy 1997). This study looks at diversity in olfactory language and culture through the example of the indigenous Chachi society of Ecuador, whose language Cha’palaa features specialized resources for talking about smell, with significant implications for debates of the supposed marginalization of olfaction. Cha’palaa bases its large set of olfactory terms on its grammaticalized “smell classifier”, a type of classifier previously unattested in the typological literature (e.g., Aikhenvald 2003). We compare Cha’palaa’s system of smell terms to that of its neighboring language Imbabura Quechua as well as to that of English, both to highlight what is distinctive about Cha’palaa, as well as to update previous accounts of the language of smell in these two other languages. We begin by looking at language structure, and then turn to language usage, drawing on natural speech corpora to examine specific cases of smell terms as used in social interaction and compare their relative discourse frequencies across the three languages. Finally, we discuss the potential significance of the linguistic facts about Cha’palaa for understanding diversity in cultures of olfaction across different societies. Cha’palaa olfactory language 177 Olfaction in Cha’palaa Speaking Society Some preliminary observations based on long-term ethnographic study in Cha’palaa speaking communities indicate that smell may have special cultural significance in Chachi society. In a tropical rainforest environment where visual perception can be limited, olfaction is a source of information about plants, animals, human activity and weather conditions that local people are acutely aware of and pause to notice and comment about.2 Olfaction is mentioned in traditional narratives such as tales of monsters that track humans by smell. References to smell are heard frequently in relation to cooking and hygiene products, and odors can play an important role in ritual, such as during the annual December Ninu~ ceremony; Floyd describes the experience: All of the surrounding communities had been gathered at the ceremonial house for days and nights of marimba and drumming that culminated in an event centrally involving perfume. Community members had explained to me ahead of time that perfume would be distributed to the crowd, but I had not realized that this meant that we would all be liberally soaked. Around midnight a group of people picked up the bottles that had been arranged on the altar, opened them, and poured perfume all over the upper body of each attendee, indicating the conclusion of the ceremonies. Fig 1. The apparent significance of olfaction in Cha’palaa daily life and ceremony was further supported by the results of a 13-language comparison of sensory verbs in natural speech corpora. In this comparison, Cha’palaa ranked the sense of smell higher relative to all other languages except for Semai (San Roque et al. 2015), one of the Aslian languages noted for their olfactory language (Tufvesson 2011). In addition, developing a full account of Cha’palaa smell terms revealed that the most frequent terms andyu “fragrance” and pudyu “stench”, are just two members of a larger set of abstract smell quality terms, all of which are formed with the morpheme -dyu, which grammatically classifies the resulting term as a type of smell. This case of a smell classifier is significant not only because it adds to the cross- linguistic counterexamples to the statement that “there is no semantic field of smells” (Sperber 1975:116), but also because the typological literature on classifiers has yet to document classifiers based on smell. The different types of classifier systems seen in the world’s languages are diverse in their grammatical expression (Corbett 1991; Senft 2000; Aikhenvald 2003), but semantically they are commonly based on core aspects like animacy, gender, shape, type of material, or usefulness (Seifart 2010:726– 727), domains which are thought to be selected due to their high “cue validity” (Aikhenvald 2003:239–240; Seifart 2010:725–726), meaning a higher probability of Figure 1. Ritual communal use of perfume (pindyupi, ‘sweetness-CL:SMELL-CL:LIQUID’)ina Chachi community 178 Journal of Linguistic Anthropology inferring the properties of an object from a cue (Beach 1964; Reed 1972; Rosch et